Assessing the Risk of Online Gambling Products: A Replication and Validation of Behavioural Markers of Harm Using the Problem Gambling Severity Index.

Abstract

In this paper, we investigate the validity of objective operator data as proxy indicators for riskier gambling as based upon an independent self-report measure. Such work is important to strengthen the validity of gambling research involving objective behavioral indicators of harm used to detect higher risk gambling or product choices. To address these aims, a total of 21,464 individual customers from a single international operator completed the Problem Gambling Severity Index (PGSI). These data were then mapped to measures of player gambling activity and a range of objective behavioural markers of harm. The results confirmed that people scoring 8 + on the PGSI were found to have higher levels of gambling involvement (participation, days active and expenditure) on a range of gambling products, with differences generally larger for casino than wagering activities. Importantly, this group was also more likely to have a higher incidence of behavioural markers of harm (e.g., declined deposits). The data allowed for the replication of a previous study using such markers to detect differences in product risk, but further validated their use in a variety of analytical contexts by showing a concordance between self-reported and objective risk measures.